The present invention relates to a liquid crystal device, and more particularly relates to a ferroelectric liquid crystal device and manufacturing method therefor.
Heretofore, it is known to utilize twisted nematic liquid crystals for devising electro-optical displays. The liquid crystal materials are employed in layer form which is finely devided into a number of pixels by virtue of a matrix electrode arrangement contiguous to the liquid crystal layer. However, due to occurence of crosstalk between adjacent pixels during operation in a time multiplexing mode, the number or densities of pixels is substantially limited.
Switching is performed by means of thin film transistors provided for every pixel, the driving fashion being a active matrix system. However, because of complexities of the manufacturing process, it is very difficult to produce a display having a large area when cost reduction is considered.
In an attempt to solve the above shortcomings of prior art devices, Clerk et al. proposed a ferroelectric liquid crystal device in their U.S. Pat. No. 4,367,924. FIG. 1 is an explanatory schematic diagram showing the action of liquid crystal molecules in the prior art devices. A ferroelectric liquid crystal is interposed between a pair of glass substrates 11 and 11' which is provided with an electrode arrangement made of In.sub.2 O.sub.3, SnO.sub.2 or ITO (Indium Tin Oxide) on the inside thereof. The liquid crystal is arranged between the substrates so that each molecular layer 12 is formed normal to the substrates as illustrated in the figure. The phase of the liquid crystal is chiral smectic C at which the device is driven, desirably at room temperature. The liquid crystal molecule takes two stable positions I and II which makes angles .theta. and -.theta. with the layer normal as shown in FIG. 2.
The position of molecules switch between the of two stable positions in the light of an externally applied electric field normal to the substrates, whereupon visual images can be constructed based on differential birefringence among pixels. One feature of this type of display device is bistablity by virtue of which the position of each liquid crystal molecule is maintained the same as the previous state even after the applied signal is removed until another signal is applied anew in the opposite sense. Namely, they can function as memory elements.
To such a ferroelectric liquid crystal device, it has been required to obtain a uniform state of liquid crystal without imperfections throughout the liquid crystal layer between a pair of substrates for uniform drive capability throughout the entire display area. The liquid crystal layer of this condition is referred to as "mono-domain" hereinafter.
Imperfections and defects are caused because of small flaws of orientation control films, unevenness of an electrode arrangement formed on the substrates, spacers and other causes. In order to avoid occurrence of such imperfections and defects, mono-domain has been developed by temperature gradient method in which crystalline structure of liquid crystal is one-dimensionally developed inwardly from one end of the display area.
However, epitaxial growth of the smectic phase from a spacer edge under appropriate temperature gradient application of the gradient temperature method is effective only when the display area of devices exceeds several squared centimeters. Furthermore, even if a large area mono-domain is constructed, the crystalline direction is not exactly aligned parallel to the substrates, making a pre-tilted angle with the substrate plane. For this reason, liquid crystal molecular layers tend to bend causing zig-zag structures. The switching due to external electric fields may take place in reverse senses at the both sides of a folding plane in the zig-zag structure. It has been often observed that uniform display and driving performance are hindered by the zig-zag structure.
The inventors have repeated experiments using liquid crystal displays comprising a chiral smectic C liquid crystal (ferroelectric liquid crystal). However, they have failed to satisfactorily drive the displays and to obtain clear images. This fail is supposed because of interaction between pixels. The main cause of interaction might be quasi-monocrystalline (homogeneously ordered without discontinuity) regions bridging adjacent pixels. In other words, the switching of one pixel might influence an adjacent pixel through the mono-crystalline region bridging therebetween.